« first day (1223 days earlier)      last day (552 days later) » 
02:00 - 13:0013:00 - 23:00

02:39
for some strange reason, the bot didn't want to announce it
so here I am doing it manually
thank you to everyone involved in developing it
and uh now back to work on the fourth one ig :p
huh
it's gonna have to be re-released
why will it need to be re-released you ask
that's why
03:22
Um
That is my bad
Frickin' windows
ysthakur pushed a commit to Vyxal/ysthakur-patch-1 in Vyxal/Vyxal: Define variables in env section of release workflow
ysthakur created branch ysthakur-patch-1 in Vyxal/Vyxal
ysthakur opened pull request #1891 (Define variables in env section of release workflow) in Vyxal/Vyxal
@lyxαl Can you force merge that without 2 reviews?
I swear on your mother that it works
(even if it doesn't, all that'll happen is people'll get slightly slower executables)
@user yes
You should be able to too
Doesn't look like it
There's a checkbox usually, I think
You don't see a button like that
03:31
No
Or something that says merge as administrator?
Nope
github-actions[bot] pushed a commit to Vyxal/ysthakur-patch-1 in Vyxal/Vyxal: Refresh element and/or trigraph docs from recent commit
Let me fix that for you rq
I think we should go back to the 1 review requirement lol
Since there's only 3 of us and the person that makes the PR can't review
Huh, when did we get CodeQL?
Ooh nice, tyvm
03:33
@user sure
ysthakur merged pull request #1891 (Define variables in env section of release workflow) in Vyxal/Vyxal
ysthakur pushed a commit to Vyxal/version-3 in Vyxal/Vyxal: Define variables in env section of release workflow (#1891)
ysthakur deleted branch ysthakur-patch-1 in Vyxal/Vyxal
Hehehe it feels so good to break the rules
Honestly I thought you already were an org admin
@user yesterday
I decided
@user done
@lyxαl Cool thanks
> There'll be a few things I've missed in both sections which user will probably call me goofy for missing, but that's something I can live with.
You didn't mention the website no longer yeeting itself
Nothing to see here
Oh one actual change: we have logging now, which is useful not only for us but also for users (though arguably not that useful for ordinary users)
 
1 hour later…
04:51
I sure do love builds taking 6 minutes :p
> [info] Optimizing (release-full mode) (371937 ms)
and that's just for windows
Yeah, first the Scala compiler (which is kinda slow) does its thing, then the Scala Native optimizer does its thing, then LLVM does its thing
release-fast and no LTO would bring the compile times down again
still waiting for the release builds :p
Oh wow it's been over 13 minutes
and it failed
...
04:59
 Error:  0  0x10b1d5f7a  __assert_rtn + 139
Error:  1  0x10b1e1793  ld::tool::SymbolTableAtom<x86_64>::addGlobal(ld::Atom const*, ld::tool::StringPoolAtom*) (.cold.3) + 35
Error:  2  0x10b0c5523  ld::tool::SymbolTableAtom<x86_64>::addGlobal(ld::Atom const*, ld::tool::StringPoolAtom*) + 1107
Error:  3  0x10b0c4b27  ld::tool::SymbolTableAtom<x86_64>::encode() + 263
Error:  4  0x10b0b847e  ___ZN2ld4tool10OutputFile20buildLINKEDITContentERNS_8InternalE_block_invoke_3 + 30
Error:  5  0x7ff8034070cc  _dispatch_call_block_and_release + 12
tripped on macos
and uh
it could be a linker bug
different error, but it mentions a failed assertion
Scala Native bug
argh
LTO Full it is,then
ysthakur pushed a commit to Vyxal/ysthakur-patch-1 in Vyxal/Vyxal: Use full LTO for MacOS builds
ysthakur created branch ysthakur-patch-1 in Vyxal/Vyxal
ysthakur opened pull request #1892 (Use full LTO for MacOS builds) in Vyxal/Vyxal
Lyxal approved #1892 (Use full LTO for MacOS builds) in Vyxal/Vyxal
merge when ready
ysthakur merged pull request #1892 (Use full LTO for MacOS builds) in Vyxal/Vyxal
ysthakur pushed a commit to Vyxal/version-3 in Vyxal/Vyxal: Use full LTO for MacOS builds (#1892)
ysthakur deleted branch ysthakur-patch-1 in Vyxal/Vyxal
I was born ready
05:09
ight 3rd rerelease
and of course I made it on the v2 branch again
there's the action
The actual compilation finishes in 2-3 minutes
it's the native stuff that makes it long, yes
Actually it was probably a bad idea to use release-full rather than release-fast
Release-full might be too aggressive
I think it's fine
the whole point of targetting native is speed
True
05:17
maybe the none option could be good
because chances are if you're using native, you're doing fastest-code
meaning you won't want a garbage collector yiking your program
Perhaps separate builds?
could be worth looking inot
*into
ysthakur pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: Add eval, frames, and stack commands to debug repl
ysthakur created branch more-debugger in Vyxal/Vyxal
gah, github actions need to not report warnings as errors
gave me a good startle
ysthakur pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: Add continue command
05:31
hey user
guess what
worst part is that is passed compilation on windows
failed on the rename :p
Oh that's the path from when it used sbt
Mill uses out/native/nativeLink.dest
ysthakur pushed a commit to Vyxal/ysthakur-patch-1 in Vyxal/Vyxal: Update path to built exe in release workflow
ysthakur created branch ysthakur-patch-1 in Vyxal/Vyxal
ysthakur opened pull request #1893 (Update path to built exe in release workflow) in Vyxal/Vyxal
ysthakur merged pull request #1893 (Update path to built exe in release workflow) in Vyxal/Vyxal
ysthakur pushed a commit to Vyxal/version-3 in Vyxal/Vyxal: Update path to built exe in release workflow (#1893)
ysthakur deleted branch ysthakur-patch-1 in Vyxal/Vyxal
06:05
ysthakur pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: feat: Add breakpoints
Let the games begin for the fourth time :p
ysthakur opened pull request #1894 (Add some stuff to the debugger) in Vyxal/Vyxal
ysthakur assigned ysthakur to pull request #1894 (Add some stuff to the debugger) in Vyxal/Vyxal
ysthakur force-pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: Add eval, frames, and stack commands to debug repl
ysthakur force-pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: Add continue command
ysthakur force-pushed a commit to Vyxal/more-debugger in Vyxal/Vyxal: feat: Add breakpoints
Gotta go to bed now, hopefully it works this time lol
o/
06:18
o/
23 minutes and still building
25 minutes now :p
07:15
I just checked and all the executables are there
It all built successfully
3-beta.3 is fully live
07:30
Btw when I click the link button on the site it doesn't copy
(Chrome for Android)
I have to click the copy to clipboard button in the output box for it to copy
Same with all the buttons, not just the link button
(@lyxαl)
Same on Samsung Internet
@TheThonnu I think that's unimplemented
Hasn't been done yet
Because it doesn't autocopy on desktop either
Should be pretty easy to do, right?
Probably
I think it just hasn't been done yet
Also, the site needs to have the dictionary loading script reverted
Because it's no longer compiled with the project
The dictionaries
They too big
Make the compiler take 10 million years
08:07
It already takes too long lol
When I want to run the tests it takes ~1 minute
I'm used to <1s with Thunno 2 in Python :p
1s tests with python?
Nonsense! Impossible! Tests don't take 1 second in python!
I don't use an external library
No dependencies, remember?
1.2s
That's 1ms per test :)
ooh I got one below 1s
How long do your tests take?
08:59
@TheThonnu a minute to collect and execute
Do note that the testing file for vyxal is like 10 times larger than yours
The main testing file that is
The other testing files obviously aren't as big
@lyxαl you have a similar number of tests though
maybe a bit more
About 500 or so
it says you have 1787 actual tests
not functions
and I have 1129
so yours would take ~2s if you did it my way :p
But then there'd be no nice trace backs and value tracking :p
Especially useful for when there's deep recursion and other things
sure
"expected x; got y" is enough for me :)
09:06
Sure it may be enough, but it sure could be a lot more helpful :p
not worth a whole minute to run tests IMO
It saves debugging time because it tells you what variables are in between function calls
fair enough
still, I want no dependencies
It could also be slow because it's testing edge cases that would otherwise take a while to compute
I'm pretty sure half of the 1s is because of my really inefficient brute-force LCM function :p
09:08
And then it also has to do two equality checks because lazylists aren't normal lists and memory references may be different
def lcm(l):
    l = [int(i) for i in l if isinstance(i, (int, float)) or str(i).isdigit()]
    if not l:
        return 0
    n = 1
    while 1:
        if all(n % j == 0 for j in l):
            return n
        n += 1
+ there's also the lexer, parser, transpiler, flags and complex tests
"Efficiency? What's that?"
And the version 3 testing framework has even more tests to consider.
Anyway, should I implement Ẹ and Ḥ?
09:12
Which are those?
partition at truthy
and head extract
just so I don't do this wrong: partition at truthy means [0,1,0,0,1,0,0,1] -> [[0],[0,0],[0,0],[]], right?
2 days ago, by lyxαl
You don't even need to ask
Alright that last question yes that's good to ask
I'd have to check what jelly does
"partition after truthy indices"
should I copy that?
That's "Partition y after truthy indices of x."
So probably after
And it'd be a dyad
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: merge conflict
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Merge branch 'version-3' of github.com/Vyxal/Vyxal into version-3
09:16
Taking two things
Not-Thonnu created branch v3-add-e-and-h-downdot in Vyxal/Vyxal
oh yeah
It also has partition before and partition without keeping borders
but they're digraphs
Yeah
I feel like I would have taken from the monographs
Gtg for now
Glhf :p
09:21
o/
09:36
@TheThonnu should we tell them about euclidean gcd
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Start to add E downdot
@lyxαl @user I got an error
[error] -- Error: /github/Vyxal/shared/src/vyxal/ListHelpers.scala:658:8 ---
[error] 658 |    for (i, j) <- makeIterable(lst).zip(makeIterable(part)) do
[error]     |        ^
[error]     |this case is unreachable since type vyxal.VAny and class Tuple2 are unrelated
Code:
  def partitionAfterTruthyIndices(lst: VAny, part: VAny)(using Context): VList =
    var res = ArrayBuffer(VList())
    for (i, j) <- makeIterable(lst).zip(makeIterable(part)) do
      res(res.length - 1) = VList.from(res(res.length - 1) :+ i.asInstanceOf[VAny])
      if j.asInstanceOf[VAny].toBool then res = res :+ VList()
    VList.from(res.toList)
oh actually I fixed it by converting to Lists
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Fix
nayakrujul opened pull request #1895 (Add E and H downdot) in Vyxal/Vyxal
github-actions[bot] pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Scalafix changes
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Update Elements.scala
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Merge branch 'v3-add-e-and-h-downdot' of github.com/Vyxal/Vyxal into v3-add-e-and-h-downdot
09:58
E downdot implemented
I'll do tests + H downdot later
github-actions[bot] pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Scalafix changes
 
2 hours later…
12:04
interesting
did we configure the bot to not report pre-releases?
cool if it's intentional
Lyxal pushed a commit to vyxal.github.io/add-the-dictionary-back in Vyxal/vyxal.github.io: Add the dictionary back
Lyxal created branch add-the-dictionary-back in Vyxal/vyxal.github.io
@TheThonnu okay so we made a zip method in VList and that conficts with the regular zip method
We could either move tgat zip to listhelpers or rename it
Tvzip or something
*to vzip
hey user when you re-added the manual way of adding dictionaries to the js, did you remember to expose the dictionary as a retrievable js file?
vyxal.github.io/Vyxal/dictionary.js 404s which is what used to be used
@user I'll rename it
Should I rename zipWith -> vzipWith as well?
Still doesn't work :(
[error] 658 |    for (i, j) <- makeIterable(lst).vzip(makeIterable(part)) do
[error]     |        ^
[error]     |this case is unreachable since type vyxal.VAny and class Tuple2 are unrelated
12:21
vZipWith
doesn't matter
it won't work
no i meant if you're renaming it that's what I recommend you name it
@lyxαl LongDuctionary.fxt and Shortzfoctionar.txt should work
@TheThonnu no use zip there
You want a regular zip
12:22
@user I need something I can use in a <script> tag :p
  <script src="https://vyxal.github.io/Vyxal/dictionary.js" async defer></script>
for example
Uhh
Maybe it was a bsd idsa to use those files
They don't get loaded lazily anyway
I need something to make the website compress strings again :p
@user still getting errors
[error] -- [E007] Type Mismatch Error: /github/Vyxal/shared/src/vyxal/Elements.scala:1936:27
[error] 1936 |        VList.from(iter.zip(ListHelpers.map(b, iter)))
[error]      |                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[error]      |Found:    Seq[(vyxal.VAny, vyxal.VNum | String | (vyxal.VFun | vyxal.VList))]
[error]      |Required: Seq[vyxal.VAny]
[error]      |
[error]      | longer explanation available when compiling with `-explain`
[error] -- [E007] Type Mismatch Error: /github/Vyxal/shared/src/vyxal/Elements.scala:1939:47
@lyxαl doesn't it already?
it's ok
We can just convert to list
12:25
@user it still assumes the dictionary is compiled with the project
it was never changed back
@TheThonnu yup, if you get errors elsewhere now, that means those places wanted the vyxal zip. Change them to use vzip and it'll work
oh come on
I just changed it back
@lyxαl ah
can you just do it after I push?
@TheThonnu not here, this still uses zip. Change the pther placew
12:27
yeah ik
The compiler errors actually show you that
It wants VAnys but your e giving it a tuple because youre using normal zip instead of vyxal zip
@TheThonnu okay I'll do it when I'm at mg paptop
nah dw I'm doing it
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Implement H downdot
@user done
Awesome
12:34
tests time
github-actions[bot] pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Refresh element and/or trigraph docs from recent commit
umm
[info] - Inputs: #["T","h","o","n","n","u"#], #[0,1,0,0,1,1#] *** FAILED ***
[info]   java.lang.Error: Lexing failed: VyxalCompilationError(Lexing failed: Expected parseAll:1:1 / end-of-input:1:1, found "Ẹ")
[info]   at vyxal.Interpreter$.execute(Interpreter.scala:15)
[info]   at vyxal.YamlTests.testFun$proxy1$1(YamlTests.scala:84)
[info]   at vyxal.YamlTests.execTests$$anonfun$4$$anonfun$1(YamlTests.scala:117)
[info]   at org.scalatest.Transformer.apply$$anonfun$1(Transformer.scala:22)
[info]   at org.scalatest.OutcomeOf.outcomeOf(OutcomeOf.scala:85)
same with the other tests
that's not the test failing, is it?
(@lyxαl @user)
"Ẹ":
  - {in: [["T", "h", "o", "n", "n", "u"], [0, 1, 0, 0, 1, 1]], out: [["T", "h"], ["o", "n", "n"], ["u"], []]}
  - {in: [[], [0, 1, 2, 3]], out: [[]]}
  - {in: [[1, 2, 3, 4], [1, 2, 3, 4]], out: [[1], [2], [3], [4], []]}
That's what's in the YAML file
that
that's certainly goofy
that ever happened before?
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Update tests.yaml
Not-Thonnu pushed a commit to Vyxal/v3-add-e-and-h-downdot in Vyxal/Vyxal: Merge branch 'v3-add-e-and-h-downdot' of github.com/Vyxal/Vyxal into v3-add-e-and-h-downdot
12:47
no
well I found the probelm
*problem
and it's uh
stupid
to say the least
me or you? :p
me
isn't in the codepage
I somehow forgot to remove it from info.txt
lol
nice one
btw why are you changing the codepage?
I'm not anymore
the v2 codepage is quite good imo
12:57
oh you mean for v3 in total
because more modifiers and I wanted more dotted letters
ᵃᵇᶜᵈᵉᶠᶢᴴᶤᶨ!ᵏᶪᵐⁿᵒᵖᴿᶳᵗᵘᵛᵂᵡᵞᶻ are all modifiers (visual consistency)
you could steal the Jelly codepage like I did :P
and be unoriginal?
haha no :p
@lyxαl I call it working smarter not harder
:P
12:59
I call it boring :p
02:00 - 13:0013:00 - 23:00

« first day (1223 days earlier)      last day (552 days later) »